INDEX

Explanations

place of residence

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 retiring

-0.12

linger

-0.11

 withdraw

-0.10

 RETURN

-0.10

 returning

-0.10

 reclaim

-0.10

 lays

-0.10

 rewind

-0.10

Return

-0.09

otta

-0.09

POSITIVE LOGITS

mor

0.10

 bare

0.09

 dressed

0.09

 Styled

0.08

 heading

0.08

 Tome

0.08

kit

0.08

/host

0.08

 ostensibly

0.08

 embarked

0.08

Activations Density 0.019%